Research On Nondestructive Detection Methods Of Clinched Joints

With the widespread use of lightweight sheets,sheet joining technology has received more and more attention.Clinching technology has huge advantages in joining lightweight sheets,and has become an important process for reducing vehicle weight and achieving energy conservation and emission reduction.However,in practical applications of clinching technology,there are often problems such as clinched joint defects and unstable strength.And the current detection of clinched joint quality is destructive,and the existing quality inspection means regarding the clinched joints can not be achieved in a real-time and non-destructive way.To solve these problems,this paper proposed the concept of the joint’s external dimension parameters for the first time,and established the mathematical relationship between the external and internal dimension parameters of joint.The mathematical relationship between the key dimension parameters of the clinched joint and the strength was deeply studied,and a non-destructive testing system that can monitor the quality of clinched joints in a real-time,non-destructive and remote way was established.By studying the mechanical properties of clinched joints under different forming forces,the mathematical relationship between the strength of clinched joint and its internal parameters was obtained.The clinched joints formed under different process parameters were studied,and two common surface defects of joints were found.The material flow mechanism of the clinched joint and generating mechanism of each defect is analyzed.The external dimension parameters of the clinched joint were defined according to the characteristics of obvious dividing line in different extension areas of the joint produced by extension die.The appearance image acquisition system of the clinched joint point was built,the visual processing method of the appearance image of the joint was deeply studied,and the feasibility of different filtering algorithms and edge detection algorithms for the appearance image processing of the joint was discussed.A vision processing solution for the identification of external dimensional parameters and defect recognition of clinched joint was proposed.The actual values of the external dimension parameters of each clinched joint are obtained.The idea of equal volume is proposed by studying the volume change law of the sheets before and after clinching process.And based on the idea of equal volume,a mathematical model between external and internal dimension parameters of the clinched joint was established.The model was verified by experimental data within the error of 3.69%(bottom thickness)and 6.43%(neck thickness).The strength calculation models of clinched joint under various failure modes were established based on load analysis and external dimension parameters of clinched joints.After verification using experimental data,the models proposed can predict the static strength and failure mode of clinched joint with the accuracy of 85.46% and 100%,respectively.Considering the large amount of calculation and high time complexity of the strength calculation model,in order to reduce the calculation time,this paper established the mapping relationship between the external dimension parameters and strength according to the characteristics that different external dimension parameters of the clinched joint produced by different process parameters are different and the corresponding static strengths are also inconsistent.By this way,the internal dimension parameters and strength of clinched joint can be predicted quickly by linear interpolation.Based on the idea of remote and real-time detection,a joint detection system(online address http://www.clinching.top)was established.The server has developed three modules: user management module,joint detection module and background management module.The user management module was used to verify whether the user has the right to enter the detection system and record the detection results of each user.The joint detection module was responsible for the detection of joint defects and strength,and saved the detection results in the database.The background management module could view all historical detection records of users,and export them in Excel format,which is convenient for users to process data.